Baubock et al. • CLIMATE CHANGE DOES NOT AFFECT PROTANDRY 
213 
TABLE 2. Relationships of mean arrival dates of males and females with climate variables as measured by reverse 
stepwise regressions that included three climate variables: mean temperature (Mean T), NAO index, and ENSO index. 
Slopes arc shown as days/ C for Mean T and days/index unit for NAO and ENSO. The P-value indicated in bold is 
significant after sequential Bonferonni correction in = 14 tests). However, seven of the 14 tests hud relationships with P < 
0.05. which is unlikely to have been caused by chance alone <x = 59,7, df = I, P < 0.001), suggesting the sequential 
Bonferroni correction may be overly conservative. 
Black-iuid-wliiu! Mourning Common American 
Warbler Warbler Yellowthroai Eastern Towhee Goldfinch Baltimore Oriole American Redstart 
F M H M F M F M F M F M F M 
Clim. var. ENSO none NAO NAO Mean T none none Mean T none none none ENSO none Mean T 
Slope -0.5 -0.6 -0.6 -1.2 -1.3 -1.0 -1.5 
/?- 0.17 0.17 0.13 0.18 0.13 0.38 0.28 
P 0.018 0.020 0.040 0,013 0.038 0.032 0.002 
within each species tended to follow similar trends 
over time and in relation to climate (Figs. 1, 2). 
No species, after correcting for multiple tests 
(n = 7), had a statistically significant relationship 
between any climate variable and extent of 
protandry, as measured by linear regression. The 
extent of protandry was not related to the strength 
of territoriality in a species, or any other 
ecological variables we examined as measured 
by ANOVA. 
Female/Male Ratio ,—Sex ratios did not change 
over time for any species (linear regression, P > 
0.05 for all species). However, after correcting for 
the number of tests (n = 7), changes in the 
female/male ratios of Mourning Warblers were 
significantly related to mean temperature in the 
year of migration with more females in years with 
higher temperature (reverse stepwise regression, 
P = 0.002) (Table 3). Variation in sex ratios 
among species was not explained by any of the 
ecological variables we tested, nor was it related 
to the average arrival date of the species or the 
average sex ratio as measured by ANOVA. 
DISCUSSION 
All species we examined demonstrated protan¬ 
dry, except Baltimore Orioles. Protandry did not 
change over time, nor did it change in response to 
any climate variables that we tested. This result 
contrasts with a previous finding that protandry 
increased in Barn Swallows ( Hinindo rustica) in 
Europe in response to climate change (Moller 
2004), but supports a study of protandry in other 
European bird species (Rainio et al. 2007). 
Our finding that protandry is common among 
the sexually dimorphic species we examined 
confirms previous results (Rubolini et al. 2004). 
However, our results do not appear to support any 
of the hypotheses explaining the development of 
protrandy in birds. Males were larger than females 
in all study species (Table 1). but there was no 
correlation between size and amount of protandry 
observed, as would be predicted by the ‘differen¬ 
tial susceptibility’ hypothesis, which suggests 
larger males may be less affected by adverse 
conditions than smaller females and can therefore 
arrive earlier (Morbey and Ydenberg 2001. Saino 
et al. 2010. Spottiswoode and Saino 2010). In 
addition, male Baltimore Orioles, which arrive 
simultaneously with females are. on average, 
larger than females similar to the other species 
we examined. Thus, it is unlikely that greater 
susceptibility to adverse conditions is preventing 
females from arriving as early as males. Further, 
males are not arriving earlier by virtue of a more 
northerly wintering area. American Redstart 
males and females, for example, winter near each 
other on Jamaica (Stpdds and Marta 2005) but 
still had significant differences in timing of 
arrival. Different wintering areas may be pari or 
all of the cause of protandry for American 
Goldfinches—a species in which males winter 
further north than females (Prescott and Mid¬ 
dleton 1990)—but this explanation does not 
appear to account for protrandy in the other 
species studied. 
Another group of hypotheses explaining pro¬ 
tandry suggest males gain a fitness advantage over 
females by arriving earlier and. thus, will arrive 
early despite non-ideal early spring conditions. 
Moller (2004) and Spottiswoode et al. (2006) 
found that more fit males arrive earlier, and that 
early arrival is correlated with improved second¬ 
ary sexual characteristics. Thus, males gain a 
reproductive advantage by arriving earlier via the 
'mate choice' hypothesis: male arrival date 
